The present disclosure relates generally to remote monitoring of industrial plants. With the advent of cloud computing systems, it is now possible for users at any network accessible location to initiate queries of any network accessible remote computing systems which are performed by a cloud computing system. Generally speaking, these techniques could be applied to a variety of computing systems including industrial plants, for example, by gather monitored or logged data from industrial plants for analysis. This holds the promise of many benefits. For example, a cloud computing system may process aggregated data pertaining to a multitude of devices located at an industrial plant to improve plant efficiency or predict maintenance of machinery or other devices in the industrial plant. Some conventional proposals are to send all data from a complete set of data sources to a centralized processor for analysis. Such approaches are bandwidth and processing resource intensive and are insecure in that all of the data pertaining to the industrial plant resources is shared with a centralized processor. Other conventional proposals request only relevant data from each relevant data source. While potentially less bandwidth and processing resource intensive, such approaches utilize a common information model stored on the cloud may be used by the cloud system for locating data sources within each industrial plant. The common information model includes information regarding the system structures as well as the data source and data source properties of the industrial plants. This poses a significant security concern for industrial data plants. Furthermore, when system structure changes, data source properties change, or data sources are added or removed, the common information model must be updated which tends to negate the initial decrease in bandwidth ad processing resource consumption. It can thus be seen that conventional proposals for remote monitoring of industrial plants suffer from a number of shortcomings and disadvantages including those respecting security of industrial plant data and infrastructure information, increasing system reliability, processing requirements, and bandwidth requirements. There remains a significant need for the unique apparatuses, methods, systems and techniques disclosed herein.